Content area
Full text
FEATURE STORAGE-AREA NETWORKING
To optimize the use of currently available high-bandwidth optical pipes, a storage designer would do well to fully understand GFP-T
The transport of storage-area networking (SAN) traffic has become a focal point of activity in the metro equipment-- design sector. Carriers are eyeing SAN as a market for increasing revenue streams as large corporations and others turn to them to help distribute SAN traffic to multiple locations or, in some cases, to centralized SAN storage facilities hanging off Sonet/Synchronous Digital Hierarchy (SDH) networks.
The real challenge carriers now face is how to handle these traffic patterns. In the past, metro equipment developers have turned to dense-wavelength division multiplexing (DWDM) techniques to carry SAN traffic over a single wavelength. The problem is that SAN traffic doesn't fully optimize these optical pipes, thereby wasting valuable bandwidth in the metro network.
In a drive to solve that problem, the International Telecommunications Union (ITU) has developed a transparent-mode version of the generic framing procedure (GFP-T) that allows designers to package SAN traffic into a virtually concatenated pipe to use fully the available bandwidth of the Sonet/SDH network to fully leverage their capabilities.
This two-part series takes an in-depth look at the transparent-mode GFP specification. In Part 1, we'll discuss how the transparent GFP method operates, how it handles encoded SAN traffic, bandwidth management, and error detection capabilities. Part 2, which can be found online now at www.commsdesign.com/csd/issue, will explore payload scrambling, client management frames and potential extensions to the transparent-mode GFP standard.
The basic GFP frame structure is shown in Fig. 1. The header uses a two-- octet-length field, which specifies the length of the GFP frame's payload area in octets, and a CRC-16 error-check code over this length field.
To synchronize alignment to this frame structure, the framer looks for a 32-bit pattern that has the proper zero cyclic redundancy checking (CRC) remainder. It then confirms that this is the correct frame alignment by verifying that another valid 32-bit sequence exists immediately following the point where the length field specifies that the current frame ends.
In GFP-T, a fixed number of client characters are mapped into a GFP frame of predetermined length, thus yielding a static payload length. This is a primary...